From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:49197) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VMhao-0006Ok-QQ for qemu-devel@nongnu.org; Thu, 19 Sep 2013 13:00:45 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1VMhag-000363-11 for qemu-devel@nongnu.org; Thu, 19 Sep 2013 13:00:38 -0400 Received: from cantor2.suse.de ([195.135.220.15]:35370 helo=mx2.suse.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VMhaX-0002y8-6l for qemu-devel@nongnu.org; Thu, 19 Sep 2013 13:00:29 -0400 Message-ID: <523B2DA2.7040406@suse.de> Date: Thu, 19 Sep 2013 19:00:18 +0200 From: =?UTF-8?B?QW5kcmVhcyBGw6RyYmVy?= MIME-Version: 1.0 References: <523B2BBF.7020801@suse.de> <523B2D55.6060008@weilnetz.de> In-Reply-To: <523B2D55.6060008@weilnetz.de> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] block/stream.c -Werror build failure List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Stefan Weil Cc: Kevin Wolf , Anthony Liguori , qemu-devel , Stefan Hajnoczi Am 19.09.2013 18:59, schrieb Stefan Weil: > Am 19.09.2013 18:52, schrieb Andreas F=C3=A4rber: >> Hello, >> >> As reported yesterday on IRC, since the last round of pulls I am getti= ng >> the following warning-treated-as-error: >> >> CC block/stream.o >> /home/andreas/QEMU/qemu/block/stream.c: In function =E2=80=98stream_ru= n=E2=80=99: >> /home/andreas/QEMU/qemu/block/stream.c:111:14: error: =E2=80=98copy=E2= =80=99 may be used >> uninitialized in this function [-Werror=3Dmaybe-uninitialized] >> cc1: all warnings being treated as errors >> make: *** [block/stream.o] Fehler 1 >> >> $ gcc --version >> gcc (SUSE Linux) 4.7.2 20130108 [gcc-4_7-branch revision 195012] >> >> However I only got this in 1 of 2 repositories I tried. It turns out >> that --enable-debug avoids this build issue. >> >> The stream.c code itself looks okay to me. In the case of ret >=3D 0, = copy >> is always assigned, including the goto code path. >> Still, could we initialize the copy variable to false or revert whatev= er >> change introduced this build regression? (-Wextra?) >> >> Regards, >> Andreas >> >=20 > This will be fixed by the next pull request for the block tree. > See also http://patchwork.ozlabs.org/patch/275503/ Just spotted it myself and added a review comment. Thanks, Andreas --=20 SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 N=C3=BCrnberg, Germany GF: Jeff Hawn, Jennifer Guild, Felix Imend=C3=B6rffer; HRB 16746 AG N=C3=BC= rnberg